about C3

about C3

Post by Daniel Parke » Sun, 06 Jan 2002 21:18:27



Is it possible to have a discussion about C3 where (i) the sceptics don't
taunt the XPers;  (ii) the XPers don't attribute all blame for the failure
of C3 on the "other"; (iii.)  participants don't feel it necessary to
respond to every post just to get the last word in; (iv) XPers don't preach
or become overly defensive; (v) participants maintain a detached attitude
and are sufficiently self aware to know when they have lost perspective, and
(vi) criticism and feedback are regarded as good things rather than bad
things?  No?  Didn't think so, judging by the other threads.  But if it were
possible, we could learn a lot.  Contrawise, nobody can learn anything from
advocacy.

For starters, other posts are suggesting that the C3 project was in its
fourth year when it was cancelled.  Four years is an awful long time to take
to build a payroll system, which quite frankly is not an R&D effort.  Two
years would be a rather long time.  Why so long?  What went on during that
period?

Regards,
Daniel Parker

 
 
 

about C3

Post by David B Lightston » Sun, 06 Jan 2002 23:31:19



Quote:> Is it possible to have a discussion about C3 where (i) the sceptics don't
> taunt the XPers;  (ii) the XPers don't attribute all blame for the failure
> of C3 on the "other"; (iii.)  participants don't feel it necessary to
> respond to every post just to get the last word in; (iv) XPers don't preach
> or become overly defensive; (v) participants maintain a detached attitude
> and are sufficiently self aware to know when they have lost perspective, and
> (vi) criticism and feedback are regarded as good things rather than bad
> things?  No?  Didn't think so, judging by the other threads.  But if it were
> possible, we could learn a lot.  Contrawise, nobody can learn anything from
> advocacy.

When you realize that there is no way to prove that a methodology works
you will conclude that advocacy is all that is available.

It is a pure marketing situation, one business definitely standing to gain by
its adoption, other standing to gain indirectly, some standing to lose, and
others simply trying to separate the meat from the bull shit

The only question is whether advocacy is occuring in good faith. Some
will express doubt about that.

<snip>

 
 
 

about C3

Post by Gerold Keefe » Mon, 07 Jan 2002 00:25:41



> When you realize that there is no way to prove that a methodology works
> you will conclude that advocacy is all that is available.

there is certainly no way to prove a methodology works, but
it is certainly possible to prove that practices work by applying
measurement.

you will find a wealth of software engineering papers that applied
measurement on software practices.

regards,

gerold

 
 
 

about C3

Post by David B Lightston » Mon, 07 Jan 2002 00:39:54




> > When you realize that there is no way to prove that a methodology works
> > you will conclude that advocacy is all that is available.

> there is certainly no way to prove a methodology works, but
> it is certainly possible to prove that practices work by applying
> measurement.

> you will find a wealth of software engineering papers that applied
> measurement on software practices.

I agree it is possible to determine the operational advantages of
various practice. Whether these measurements have been performed
for XP remains to be established

Additionally selection of practices implies the consideration of
a number of tradeoffs. There is no reason to believe that the
integration of a number of good practices will yield a good
benificial synergism (ie assume the integration activity is non-linear
and that superposition does not apply)

If you have difficulty in accepting that non-linear integration can
yield problems consider chemistry.

Ammonia is a good cleaner
Bleach (I believe thats HydroClorous Acid) is a good dissinfectant

The mixture is a poisionous gas, not useful as a cleaning agent,
not really useful as a dissinfectant (It may have other uses)

Quote:

> regards,

> gerold

 
 
 

about C3

Post by Daniel Parke » Mon, 07 Jan 2002 03:22:36



Quote:> Re. 1# C3 spent most of it's time in non-XP development. During this time
it
> made no significant progress. It spent c.1yr in XP mode, during which time
> it reached 66% of the goals set.

Fair enough, this may not be a good example then.

Quote:> There was a rather lot in the XP books about how complicated the C3
> payroll system actually is. There is a reason that there were x number
> of classes and y number of methods (and z number of acceptance tests and
> w number of unit tests).

Yes, but that's true generally about non trivial projects - they are
complicated.  There are far more complicated systems out there than payroll!
Somehow people have managed to build them before.

Quote:> How about looking at this XP project:

> <http://www.xpuniverse.com/XPU03.pdf>

> Steering the Car:
> Lessons Learned from an Outsourced XP Project
> Natraj Kini
> Steve Collins

Thanks.

What I was hoping to do was to raise a discussion of a fair sized
traditional project like payroll done in XP style, and where some of the
people that worked on the project were available as resources on this
newsgroup, to respond to questions.    I'd like to see case studies and in
order to relate to the case studies I'd like to see some traditional
metrics, like number of screens, number of reports, number of database
tables, and number of batch reports, to get a sense of scale.

Regards,
Daniel

 
 
 

about C3

Post by Martin Fowl » Mon, 07 Jan 2002 03:50:24


On Sat, 5 Jan 2002 07:18:27 -0500, "Daniel Parker"


>Is it possible to have a discussion about C3 where (i) the sceptics don't
>taunt the XPers;  (ii) the XPers don't attribute all blame for the failure
>of C3 on the "other"; (iii.)  participants don't feel it necessary to
>respond to every post just to get the last word in; (iv) XPers don't preach
>or become overly defensive; (v) participants maintain a detached attitude
>and are sufficiently self aware to know when they have lost perspective, and
>(vi) criticism and feedback are regarded as good things rather than bad
>things?  No?  Didn't think so, judging by the other threads.  But if it were
>possible, we could learn a lot.  Contrawise, nobody can learn anything from
>advocacy.

The problem is that it takes time to understand why projects get into
trouble, and it usually isn't for a single, simple reason. In my
consulting days I would often get involved in projects in trouble, and
it takes a while to really understand what's going on - involving
talking to all the stakeholders. Although I know quite a few people
connected with C3, there are too many gaps in my knowledge to really
say anything worth saying. (And I have more sources of information
than most of the people passing judgement on this list.)

In any case the time when C3 was important has long passed. XP has
moved on a long way since then, and there many more project
experiences to chew on (a recent IEEE Software had three such
projects.)

Martin

 
 
 

about C3

Post by Gerold Keefe » Mon, 07 Jan 2002 04:22:43



> In any case the time when C3 was important has long passed. XP has
> moved on a long way since then, and there many more project
> experiences to chew on (a recent IEEE Software had three such
> projects.)

martin, just for the record, you could have mentioned that
the first was by James Grenning of objectmentor,
the second was by Peter  Schuh of your company, thoughtworks,
and the third was Charles Poole and Jan Willem Huisman of iona on
a project that at iona with the involvement of kent beck.

to the best of my knowledge one of the three projects was cancelled,
as was C3, and the Poole/Huisman project is maintenance.

also for the record, there is one article by Paulk (SEI) on CMM/XP.

the reason for the cross posting is to inform a wider community
that IEEE publications seem to be entered as a marketing plattform
for XP. with four XP articles in the IEEE Software and on in
the IEEE Computer in the recent issues only.

regards,

gerold

 
 
 

about C3

Post by Jordan Bort » Mon, 07 Jan 2002 04:26:41



> On Sat, 5 Jan 2002 07:18:27 -0500, "Daniel Parker"

<snip>
> In any case the time when C3 was important has long passed. XP has
> moved on a long way since then, and there many more project
> experiences to chew on (a recent IEEE Software had three such
> projects.)

> Martin

Hello Martin:

 It looks like your input might have gotten truncated: "a recent IEEE
Software had three such
projects."

 Is this a magazine you are referring to?

 Can you cite for us what you consider the largest/longest corporate
projects using XP (possibly the article you mentioned) as well as URL or
journal/date information?

 I for one could care less about C3; what I am really interested is real
world metrics based on real world projects.

 C3 has been the "legendary" metric to date, however fresh data is the most
useful.

 Regards,
 Jordan